27.10.2014 Views

Russel-Research-Method-in-Anthropology

Russel-Research-Method-in-Anthropology

Russel-Research-Method-in-Anthropology

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Structured Interview<strong>in</strong>g I: Questionnaires 285<br />

the survey away. If you send them just the right amount, they are likely to fill<br />

out the survey and return it.<br />

Warr<strong>in</strong>er et al. (1996) confirmed this f<strong>in</strong>d<strong>in</strong>g. They offered people <strong>in</strong><br />

Ontario, Canada, $2, $5, or $10 to send back a mailed survey. Factor<strong>in</strong>g <strong>in</strong> all<br />

the costs of follow-up letters, the $5 <strong>in</strong>centive produced the most returns for<br />

the money.<br />

First-class postage and monetary <strong>in</strong>centives may seem expensive, but they<br />

are cost effective because they <strong>in</strong>crease the response rate. Whenever you th<strong>in</strong>k<br />

about cutt<strong>in</strong>g corners <strong>in</strong> a survey, remember that all your work <strong>in</strong> design<strong>in</strong>g a<br />

representative sample goes for noth<strong>in</strong>g if your response rate is low. Random<br />

samples cease to be representative unless the people <strong>in</strong> it respond. Also<br />

remember that small monetary <strong>in</strong>centives may be <strong>in</strong>sult<strong>in</strong>g to some people.<br />

This is a cultural and socioeconomic class variable that only you can evaluate<br />

<strong>in</strong> your specific research situation. (See Church [1993] for a meta-analysis of<br />

the effect of <strong>in</strong>ducements on response rates to mailed surveys.)<br />

9. Contact and follow-up: Pay careful attention to contact procedures. Send a letter<br />

to each respondent expla<strong>in</strong><strong>in</strong>g the survey and <strong>in</strong>form<strong>in</strong>g the respondent that a<br />

questionnaire will be com<strong>in</strong>g along soon. Send a postcard rem<strong>in</strong>der to all potential<br />

respondents a week after send<strong>in</strong>g out the questionnaire. Don’t wait until the<br />

response rate drops before send<strong>in</strong>g out rem<strong>in</strong>ders. Some people hold onto a questionnaire<br />

for a while before decid<strong>in</strong>g to fill it out or throw it away. A rem<strong>in</strong>der<br />

after 1 week stimulates response among this segment of respondents.<br />

Send a second cover letter and questionnaire to everyone who has not<br />

responded 2 weeks later. F<strong>in</strong>ally, 4 weeks later, send another cover letter and<br />

questionnaire, along with an additional note expla<strong>in</strong><strong>in</strong>g that you have not yet<br />

received the respondent’s questionnaire, and stat<strong>in</strong>g how important it is that<br />

the respondent participate <strong>in</strong> the study. Heberle<strong>in</strong> and Baumgartner (1978,<br />

1981) found that send<strong>in</strong>g a second copy of the questionnaire <strong>in</strong>creases<br />

response rate 1%–9%. As there does not appear to be any way to predict<br />

whether the <strong>in</strong>crease will be 1% or 9%, the best bet is to send the extra questionnaire.<br />

When you send out the second copy of the questionnaire, send the packet<br />

by certified mail. House et al. (1977) showed that certified mail made a big<br />

difference <strong>in</strong> return rate for the second follow-up.<br />

Does All This Really Make a Difference?<br />

Thurman et al. (1993) were <strong>in</strong>terested <strong>in</strong> the attitudes and self-reported<br />

behaviors of people who admit to drunken driv<strong>in</strong>g. Us<strong>in</strong>g Dillman’s TDM,

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!